Fast federated machine unlearning with nonlinear functional theory

T Che, Y Zhou, Z Zhang, L Lyu, J Liu… - International …, 2023 - proceedings.mlr.press
Federated machine unlearning (FMU) aims to remove the influence of a specified subset of
training data upon request from a trained federated learning model. Despite achieving …

Accelerated federated learning with decoupled adaptive optimization

J **, J Ren, Y Zhou, L Lyu, J Liu… - … on Machine Learning, 2022 - proceedings.mlr.press
The federated learning (FL) framework enables edge clients to collaboratively learn a
shared inference model while kee** privacy of training data on clients. Recently, many …

Fedasmu: Efficient asynchronous federated learning with dynamic staleness-aware model update

J Liu, J Jia, T Che, C Huo, J Ren, Y Zhou… - Proceedings of the …, 2024 - ojs.aaai.org
As a promising approach to deal with distributed data, Federated Learning (FL) achieves
major advancements in recent years. FL enables collaborative model training by exploiting …

Expressive 1-lipschitz neural networks for robust multiple graph learning against adversarial attacks

X Zhao, Z Zhang, Z Zhang, L Wu, J **… - International …, 2021 - proceedings.mlr.press
Recent findings have shown multiple graph learning models, such as graph classification
and graph matching, are highly vulnerable to adversarial attacks, ie small input …

Federated fingerprint learning with heterogeneous architectures

T Che, Z Zhang, Y Zhou, X Zhao, J Liu… - … conference on data …, 2022 - ieeexplore.ieee.org
Recent studies on federated learning (FL) have sought to solve the system heterogeneity
issue by designing customized local models for different clients. However, public dataset …

Aedfl: efficient asynchronous decentralized federated learning with heterogeneous devices

J Liu, T Che, Y Zhou, R **, H Dai, D Dou… - Proceedings of the 2024 …, 2024 - SIAM
Federated Learning (FL) has achieved significant achievements recently, enabling
collaborative model training on distributed data over edge devices. Iterative gradient or …

Integrated defense for resilient graph matching

J Ren, Z Zhang, J **, X Zhao, S Wu… - International …, 2021 - proceedings.mlr.press
A recent study has shown that graph matching models are vulnerable to adversarial
manipulation of their input which is intended to cause a mismatching. Nevertheless, there is …

Unsupervised adversarial network alignment with reinforcement learning

Y Zhou, J Ren, R **, Z Zhang, J Zheng… - ACM Transactions on …, 2021 - dl.acm.org
Network alignment, which aims at learning a matching between the same entities across
multiple information networks, often suffers challenges from feature inconsistency, high …

Adversarial attack against cross-lingual knowledge graph alignment

Z Zhang, Z Zhang, Y Zhou, L Wu, S Wu… - Proceedings of the …, 2021 - aclanthology.org
Recent literatures have shown that knowledge graph (KG) learning models are highly
vulnerable to adversarial attacks. However, there is still a paucity of vulnerability analyses of …

Robust network alignment via attack signal scaling and adversarial perturbation elimination

Y Zhou, Z Zhang, S Wu, V Sheng, X Han… - Proceedings of the Web …, 2021 - dl.acm.org
Recent studies have shown that graph learning models are highly vulnerable to adversarial
attacks, and network alignment methods are no exception. How to enhance the robustness …